© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 5400
    Prev
    Next
    INDEX
    Explanations

    damage or injury

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĤłä¹ħ
    -0.10
     lod
    -0.10
    esson
    -0.10
    åĽº
    -0.09
     маÑĤеÑĢи
    -0.09
    æľºèĥ½
    -0.09
    ç¹ģ
    -0.08
    åĮĸåIJĪ
    -0.08
    зав
    -0.08
    æµ·éĩı
    -0.08
    POSITIVE LOGITS
    åĪĴ
    0.12
    æĪ³
    0.10
    åĬĥ
    0.10
    å¨ĩ
    0.09
    çļĦä¸ľè¥¿
    0.09
    æīī
    0.09
    çīµ
    0.09
    éªĬ
    0.08
    ther
    0.08
    åħ¥åħļ
    0.08
    Activations Density 0.070%

    No Known Activations